Project 1 will test the hypothesis that poststroke dementia is a function of preclinical Alzheimer's disease (AD). The prediction from this hypothesis is that poststroke dementia will be disproportionately represented in stroke individuals with preclinical AD in comparison with stroke individuals without preclinical AD. The Specific Aims of the project are to: 1. Over 3.5 years, enroll a cohort of 240 older adults, age 65 years and older, hospitalized with acute ischemic stroke and obtain baseline demographic data, medical history, neurological status, neuropsychological evaluation, prestroke cognitive and functional status, and blood for Project 2 (plasma) and Project 4 (DMA). 2. In this hospitalized cohort, obtain magnetic resonance imaging (MRI) to characterize cerebral ischemic lesions and positron emission tomography (PET) with the [11C]benzothiazole amyloid tracer, Pittsburgh Compound B (PIB), to determine the presence or absence of preclinical AD (as defined by a positive PIB scan). 3. Every three months after discharge, maintain telephone contact with the patient and their families to monitor intercurrent events and maintain compliance. 4. One year poststroke, assess all patients with the Clinical Core clinical and neuropsychological batteries and derive a Clinical Dementia Rating (CDR) for correlation with baseline PIB status (CDR > 0.5 will be considered evidence for dementia). 5. Obtain MRI one year poststroke to evaluate potential new ischemic lesions; obtain blood for plasma biomarkers (Project 2); enroll patients into Project 3; and follow all patients annually with clinical and cognitive assessments and obtain voluntary consent for autopsy. This project directly addresses the overall aim of the program project grant, to characterize preclinical AD. It further addresses the unresolved issue of why dementia occurs in some, but not all, individuals with stroke. This project uses resources from all Cores. It also interacts with each of the individual projects